1. Field of the Invention
The present invention relates to a solid-state image pickup device and an image pickup system, and in particular, to a CMOS type solid-state image pickup device and a camera, which are adapted for both static image pickup and moving image pickup.
2. Description of the Related Art
In recent years, with being characterized in that a CMOS type solid-state image pickup device is low in power consumption and easy to be accelerated in comparison with a CCD (Charge Coupled Device), the CMOS type solid-state image pickup device begins to spread quickly as an image sensor for a portable device, or for a single-lens reflex camera which perform high-speed rapid shooting.
In order for a CMOS type solid-state image pickup device to spread further, the engineering development which can perform still image pickup and moving image pickup in high definition is required.
Heretofore, in a MOS type solid-state image pickup device, a method of performing high-definition static image pickup and moving image pickup using analog memory cell provided in an image pickup device is described in Japanese Patent Application Laid-Open No. H02-65380 which is a Japanese patent. According to this invention, in the MOS type solid-state image pickup device, the following disclosure is performed as a method of suppressing the blurring of an image which the time difference of image pickup between an even number field and an odd number field generates. Analog memory cell besides a pixel unit is provided, and a signal obtained by photoelectric conversion in the pixel unit is once transferred per line to memory means. Then, after that, the timing difference of charge storage between the odd number field and even number field is made small by transferring a store signal of the memory means concerned by turns to an output signal line. In this invention, memory is used in order to enhance image quality at the time of still image pickup, and memory has the number of bits corresponding to full pixels. On the other hand, it is made that moving image pickup is performed at a video rate without going through memory.
It becomes hard to perform readout at the video rate at the time of moving image pickup as the pixel count of a still image increases. For example, supposing the pixel count of 3 million pixels or more is necessary in high-definition static image pickup, it is necessary to perform readout at speed of 90 MHz so as to perform readout at intervals of 1/30 sec. Then, a method of reducing the pixel count of moving images with the pixel count of a still image being maintained has been studied. One method of them is pixel addition, that is, a method of adding a plurality of photoelectric conversion signals, and decreasing the pixel count to be read.
An example of the pixel addition method in a CMOS type solid-state image pickup device is described in Japanese Patent Application Laid-Open No. H09-247535 which is a Japanese patent. What is described in this invention is a method of the memory consisting of a switching device and a capacitative element for every readout line from pixels is provided for multiple lines, and outputting an average of a plurality of stored signals by turning on concurrently switching devices of a plurality of stored signals to be added. In this way, in the CMOS type solid-state image pickup device, “pixel addition” is equivalent to “pixel averaging.” In this invention, an effect of decreasing random noise by averaging read signals after reading the same pixel signal multiple times is described. By using the same structure, it becomes possible to read the signal in the pixel count, which is smaller than the case that an unnecessary signal is skipped by averaging other pixel signals, in low random noise and low fixed pattern noise. Owing to this method, in comparison with the case of no addition, both of the random noise and fixed pattern noise become:1/√(added pixel count)For example, when the addition of 2×2=4 pixels is performed, random noise and fixed pattern noise becomes:1/√4=½in comparison with the former noise.
In addition, a solid-state image pickup device which is provided with an amplifier with a gain exceeding 1 in each column is disclosed in Japanese Patent Application No. 2003-51989 which is a Japanese patent.
In view of the above-mentioned conventional examples, what is conceivable is the structure of reading only the pixel count necessary for using addition means when the pixel count increases in an image pickup apparatus where analog memory cell is mounted. In that case, the following subjects arise in extension of conventional inventions.
(1) When the pixel count increases, the readout through memory becomes more important in moving images rather than in a still image. In moving images, there also exist restrictions of formats in an output device, and large pixel count is unnecessary. The new structure that a still image and moving images are imaged in high definition with a mega-pixel-class image pickup apparatus is necessary.
(2) Since the pixel addition, that is, the averaged readout makes the random noise of pixels small, an S/N improves. Nevertheless, in CCD technology, not only random noise becomes small, but also a signal component becomes large. Thus, in the CCD, a charge number is added by add operation and a value of an S of the S/N also becomes large. Owing to this, in consideration of a CCD, new means of improving the S/N in the readout method of adopting the add operation is demanded in a CMOS type solid-state image pickup device.